Fast Lock Tracking bypasses the compromises of other insect tracking techniques. (A) Conceptual view of tracking systems based on fixed cameras, radar, and Fast Lock-On (FLO) tracking. (B) In the starting configuration (i), the system is oriented such that the reflector carried by the insects is reflected in the center of the image sensor. When the insect moves, the reflected spot is initially offset from the center of the sensor (ii). Based on image processing and motor action, the system automatically adjusts the mirror position to recenter the spot with low latency (iii). (C) Block diagram showing that the input to the image sensor is the angular difference between the insect angle and the motor-driven optical path angle. (D) Examples of insects carrying representative retroreflective markers tracked with FLO systems include the honeybee Apis mellifera (marker diameter 3 mm, mass 20 mg) and the grasshopper Schistocerca gregaria (marker mass 70 mg). Credit: bioRxiv (2023). DOI: 10.1101/2023.12.20.572558
Our ability to learn more about insect behavior, which affects ecology, health and the economy on a global scale, depends largely on appropriate recording technology. But until now, these tools have been significantly limited.
To address this problem, a research team from Albert-Ludwigs-Universität Freiburg in Germany developed an adaptable system that can be combined with different types of hardware to produce high-speed video recordings of insect flight behavior and follow, over significant distances. , their trajectories in nature.
The team describes this technology in a paper titled “High-Resolution Outdoor Insect Videography Using Fast-Lock Tracking,” published on the bioRxiv preprint server. This new work follows their previous research on this topic, published in 2020.
Existing recording methods and their limitations
Historically, insect behavior researchers have relied on direct observation to gain new insights, and more recently, they have used harmonic radar tracking to gain information about insect flight paths, particularly those of insects. bees. The resolution of this type of radar is however limited both spatially and temporally.
In laboratories, researchers have used higher-resolution cameras to study the details of insect flight behavior, but the usefulness of these cameras does not extend to insects’ natural environments, where a number of variables can play in flight function.
Researchers have also used fixed-camera videography (in which the camera can automatically track the subjects’ movements) with some success, but this method is used most effectively with a limited number of pixels for a specific width of field. Applying this method to a wider field results in lower angular resolution, while adding pixels blurs the subject’s movement. It was not possible to magnify the subject while manually and continually adjusting the camera’s aim to track movement.
Meanwhile, the combination of high-magnification optics and high-speed image tracking works with larger subjects such as birds, drones and sports balls, but low latency is needed for smaller ones. subjects like insects. Surrounding flying insects with multiple cameras or using simplified indoor backgrounds can help, but these methods also have obvious limitations.
However, retroreflectors represent a potential solution for efficient and detailed insect tracking. Informed by studies using a previously developed method for tracking insects, the research team behind this new work has dubbed the Fast Lock-On (FLO) Tracking Solution and reports on its efforts to use it to follow the flight of insects outside.
FLO monitoring with insects
To use FLO tracking with flying insects, a tiny, lightweight retroreflective marker is affixed to a subject. When the subject moves, the marker sends a signal to an optical sensor, orienting the sensor’s optical axis while minimizing the subject’s divergence from the sensor’s center point. In this study, the researchers integrated infrared illumination near the optical axis of the sensor, and oriented and tilted the optical axis to recreate the flight paths of their test subjects, including grasshoppers (Schistocerca gregaria), bumblebees (Bombus terrestris) and domestic bees. (Apis mellifera) – via the angles detected by the FLO system.
As their paper explains, “Low latency between sensor input and motor output will improve system performance and enable higher magnification optics, which ultimately can lead to additional performance gains . From a control point of view, FLO is a closed-loop design in which the displacement of the target image originating from the center of the sensor is an error angle between the angle of the insect and the angular position of the optical path under control.
The team’s test videos successfully captured the test subject’s flight “from takeoff to landing with high magnification and low motion blur,” and the insects’ appendages remained in focus during flight.
Additionally, the system is inexpensive to create and can be simply built with a computer, a low-latency digital camera, and a pan-tilt motor system. The article describes in detail the hardware, calculations for angle reconstructions and implementation methods.
Large diameter optical path shared with a high-speed telescopic video camera to record high-resolution video of insects during their natural behavior. (A) A large diameter optical path is driven by a Fast Lock-On system integrating active infrared illumination and a pair of low-latency infrared stereo cameras. A wavelength-selective mirror allows high-speed videography of the scene along the same optical axis. Stereo disparity is used to control a focus motor on a high magnification lens. (B) The Fast Lock-On core uses one camera from a pair of stereo cameras and paraxial infrared illumination to provide information about the position of a small retroreflective marker which is then used to drive a pair of motors to reorient the optical axis over a large angular range. (C) Example image from a high-speed, high-resolution video of a toy quadcopter. (D) Example frames from a high-speed video of a bumblebee landing. Marker size: 3 mm in diameter, mass: 20 mg. Credit: bioRxiv (2023). DOI: 10.1101/2023.12.20.572558
Opportunities for system improvement
The team recognizes that tracking insect flights outdoors is an important problem that has not yet been fully solved. At least three major weaknesses remain. First, a momentary loss of tracking within the FLO closed-loop system could lead to complete tracking failure.
Additionally, “the system does not directly estimate the distance to the subject,” the document states. Finally, while precise calibration is not necessary for a successful tracking process, extracting and accurately determining the final 3D coordinates will require it.
Suggested improvements include using multiple FLO systems working together for improved computation via triangulation and eliminating the need for retroreflectors through the development of advanced computer vision techniques to detect objects on insects.
However, researchers remain optimistic that FLO tracking could be used to study many aspects of insect flight behavior, including several that are directly linked to broader ecological issues, such as insect responses to artificial lights. nocturnal, loss of habitats and pesticide treatments in their feeding areas.
More information:
T. Thang Vo-Doan et al, High-resolution outdoor insect videography using fast-lock tracking, bioRxiv (2023). DOI: 10.1101/2023.12.20.572558
© 2024 Science X Network
Quote: New high-speed video system uses sensors and infrared lighting to track insects in vast wilderness areas (January 2, 2024) retrieved January 2, 2024 from
This document is subject to copyright. Except for fair use for private study or research purposes, no part may be reproduced without written permission. The content is provided for information only.